Author: Omid David Tabibi
Date: 15:34:56 11/17/03
Go up one level in this thread
On November 17, 2003 at 18:08:03, Amir Ban wrote:
>On November 17, 2003 at 11:56:51, Jorge Pichard wrote:
>
>>On November 17, 2003 at 08:13:26, Omid David Tabibi wrote:
>>
>>>On November 17, 2003 at 07:58:59, Jorge Pichard wrote:
>>>
>>>>I kept reading from Mig that not too many programs would have known what plan to
>>>>make, such as moving the Knight from f6 and advancing the f pawn down. Most
>>>>human players, even weak players know that if your opponent has the center lock
>>>>and is advancing pawns or playing on one side of the board, your best plan would
>>>>be to play on the other side of the board, not with your pieces but with your
>>>>pawns. My question to this forum is how many other top programs would have
>>>>chosen the correct plan?
>>>>
>>>
>>>[D]rnb2rk1/1p1nbppp/1Pp1q3/N1Pp4/1B1Pp3/P1N1P2P/1R3PP1/3QKB1R b K - 0 1
>>>
>>>Kasparov,G - X3D Fritz
>>>rnb2rk1/1p1nbppp/1Pp1q3/N1Pp4/1B1Pp3/P1N1P2P/1R3PP1/3QKB1R b K - 0 1
>>>
>>>Analysis by Falcon:
>>>
>>>20...Na6 21.Bxa6 Rxa6 22.0-0
>>> ² (0.69) Depth: 3/7 00:00:00
>>>20...Na6 21.Bxa6 Rxa6 22.0-0 Bh4
>>> ² (0.59) Depth: 4/9 00:00:00
>>>20...Na6 21.Bxa6 Rxa6 22.0-0 Qf5 23.Qg4
>>> ± (0.76) Depth: 5/10 00:00:00
>>>20...f5 21.Qh5 Nf6 22.Qg5 Nbd7
>>> ± (0.73) Depth: 5/23 00:00:00
>>>20...f5 21.Nb3 f4 22.Qg4 Qh6 23.Kd2 fxe3+ 24.fxe3
>>> ² (0.60) Depth: 6/23 00:00:00 27kN
>>>20...f5 21.Be2 f4 22.exf4 Rxf4 23.Bg4 Qf7 24.0-0
>>> ± (0.74) Depth: 7/23 00:00:00 49kN
>>>20...Nf6 21.f4 exf3 22.Qxf3 Re8 23.Bd3 Nbd7
>>> ± (0.72) Depth: 7/23 00:00:00 97kN
>>>20...Nf6 21.Ne2 Na6 22.Nf4 Qd7 23.Bc3 g5 24.Nh5
>>> ± (0.83) Depth: 8/23 00:00:01 167kN
>>>20...f5 21.Ne2 g5 22.Nc3 Na6 23.Bxa6 Rxa6 24.0-0 f4
>>> ² (0.61) Depth: 8/23 00:00:02 311kN
>>>20...f5 21.Be2 Na6 22.0-0 Nxb4 23.axb4 Bh4 24.Qd2 Qg6 25.g3
>>> ± (0.74) Depth: 9/24 00:00:03 568kN
>>>20...f5 21.Be2 Na6 22.0-0 Qg6 23.Qd2 Bh4 24.Ra2 Nxb4 25.axb4 Re8
>>> ² (0.69) Depth: 10/28 00:00:08 1328kN
>>>20...Na6 21.Bxa6 Rxa6 22.0-0 f5 23.f4 Qg6 24.g4 Qh6 25.Rh2 Qe6
>>> ² (0.67) Depth: 10/28 00:00:09 1451kN
>>>20...Na6 21.Bxa6 Rxa6 22.0-0 f5 23.Ne2 Nf6 24.Qc2 Nh5 25.Rbb1 Qg6 26.g3
>>> ± (0.74) Depth: 11/28 00:00:12 1989kN
>>>20...f5 21.Be2 Bg5 22.h4 Be7 23.Qd2 Na6 24.Ra2 Re8 25.Qc2 Nxb4 26.axb4
>>> ± (0.73) Depth: 11/31 00:00:19 2964kN
>>>
>>>(David Tabibi, MyTown 17.11.2003)
>>>PIII/733MHz
>>>
>>
>>
>>Thanks for replying, I still think that at move 20...f5 was not too late to
>>save the game!
>>
>>Jorge
>
>Too late ? You must be joking. Black could play it any time in the next 15
>moves.
>
>I don't understand this fixation on f5. What's wrong with any other pawn advance
>? Black can play h5, g5, f5, and take it from there, any time during the game.
The key move was f7-f5-f4, possibly supported by g5. I don't see any point in a
move such as h5 (although Falcon liked it at several points), as the plan must
be attacking the base of the pawn chain.
>What is white going to do about it ?
What is black going to do without f5 ?
White doesn't have to be even slightly worried about any plan not containing
f5-f4.
>
>I think this game was a low swindle by Kasparov. He played a dubious line based
>on preparation that showed Fritz will not punish him for it. He gave Fritz about
>20 free tempi (5 moves to capture the pawn, 5 to disentangle, 5 to move his king
>and 5 to break).
In closed positions sometimes you can spend 10 tempi just to slightly improve
the position of a piece. Not even in closed positions actually, in many lines of
Sicilian you see frequent wastes of tempi by black (e.g., Bc8-d7-c8 in some
Scheveningen lines).
As long as black didn't opt for f5, white could have wasted as many tempi as he
wished, without worrying about anything.
>
>This was not anti-computer strategy. Kasparov played computer strategy! Locking
>the position with a horribly cramped reverse chain, taking a worthless pawn
Worthless pawn?! After you teach Junior how to count pawns, you might no longer
think so :)
The pawn was far from worthless. Kasparov "horribly" locked the position just to
grab that pawn (by playing b6). Without that pawn there was no way for Fritz to
stop white from getting a supported passed pawn, which determined the game.
>and
>hoping the sky will not fall on its head is just what a computer is expected to
>do.
Winning a game by grabbing a pawn is not exactly Junior's style, I know. But it
was the correct strategy in that position, which was perfectly executed by
Kasparov.
I would rephrase Mikhail Tal's famous statement ("Fischer is Fischer, but a
knight is a knight") as "Fritz is Fritz, but a pawn is a pawn"!
>
>Amir
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.